CHARACTERIZING THE NASOPHARYNGEAL MICROBIOME, RESISTOME, AND VIRULOME OF PREWENAED DAIRY CALVES FOR THE IDENTIFICATION OF BRD BIOMARKERS

<b>Project Methods</b><br> Study Design and sampling: a case-control study design will be used, where individual preweaned calves (up to 7 weeks of age) with BRD will be matched with controls and sampled only once using a DNS approach, as previously described14. Samples will be collected for 12 months through monthly sampling visits, as season has been shown to be an important risk factor for BRD15. Case definition and protocol for enrollment of calves with acute BRD follow: 1) initial identification of BRD suspect cases using a validated visual clinical scoring system for BRD16; and 2) confirmation either through lung auscultation or thoracic ultrasound findings, as previously described17. For each diseased calf with BRD sampled (DC), a healthy control calf without BRD (HC) or other visual signs of disease or record of treatment with antimicrobial treatment will be enrolled.Sample collection and processing: Deep nasopharyngeal swabs (DNS) culture were chosen as the sample of choice for this study because DNS culture results have been shown to have a strong agreement with transtracheal wash (TTW) and bronchoalveolar lavage (BAL) in calves clinically affected by BRD14. Advantages of DNS over TTW and BAL include: less invasive, lower cost, and requires a lower level of technical skills, increasing its potential for implementation as a routine diagnostic test in livestock production settings. For our study, two DNS will be collected from each calf enrolled14, with one being placed in transport media, and another in a sterile RNASE/Dnase free tube. Both samples will be stored in a cooler with ice and transported to UC Davis. Upon arrival, a sample will be submitted to the California Animal Health & Food Safety (CAFHS) Laboratory for aerobic culture, speciation, and antimicrobial susceptibility testing; and sample for DNA extraction will be stored at -80ºC.Sample size: Base on previously published data, the dominant BRD pathogens expected for aerobic culture will be Manhemia haemolytica (MH) and Pasteurella multocida (PM), with the low-end culture recovery rates reported at 21 and 36%, respectively. Based on this information, we will collect samples from 150 calves in the DC, securing an estimated minimum of 36 and 65 samples from calves with BRD per year per DNS sample with culture positives results for MH and PM, respectively; our goal will be to sample ~15 samples from calves in each DC and HC during each monthly sampling visit. Due to budget limitation, for sequencing, a sub-sample of samples will be selected for microbiota characterization, with: 50 samples from calves in the DC group with culture positive aerobic results and 12 samples from the DC group with culture negative aerobic results; for both cases, samples will be randomly selected and stratified by season. Furthermore, the 62 samples from the HC group that match the DC samples selected will also be sequenced as a control group. A standardized approach for sample size calculation for metagenomic approaches has not been established due to various factors, but based on previous studies, we expect that a total of 124 animals per treatment group will suffice18,19Sample Processing and DNA Extraction: DNA extraction from swabs, library preparation, sample purification, and normalization will be conducted as previously described20,21.Next-Generation Shotgun Metagenomic Sequencing: DNA fragments will be paired-end sequenced (150 base pair reads) on the Illumina NovaSeq 6000 system at the UC Davis DNA Technologies & Expression Analysis Core Laboratory. Data will be quality-filtered and aligned by the Core facility, and uploaded, processed for quality control, and annotated in the Metagenome Rapid Annotation using Subsystem Technology (MG-RAST) server21.Data analysis: Define agreement between culture-based and metagenomic data: shortly, define an optimal discrimination point for relative abundance for each BRD pathogen data using a receiver operator characteristic (ROC) analysis and compare results with culture based for standard accuracy measures (e.g. sensitivity, specificity); Microbial composition, resistome and virulome, and detection of biomarkers: Analyze the relative abundances of different microbial taxa and functions as covariates in stepwise discriminant analysis models differentiate samples from DC and HC groups. Use PathoFact, a recently developed pipeline for the identification of virulence factors, bacterial toxins, and antimicrobial resistance genes in metagenomic data, to identify potential differences between DC and HC groups 22. Use the linear discriminant analysis (LDA) effect size (LEfSe) method for the detection of biomarkers in the metagenomics data of DG samples 23,24.